The role of extracellular calcium in pregnancy-induced attenuation of phenylephrine contraction in rat aorta with functional endothelium

The effect of pregnancy on the supply of calcium ions for the contractile responses of rat aortic rings to phenylephrine was investigated. The contractility of intact aortic rings from pregnant rats, compared with that of similar rings from non-pregnant rats, to phenylephrine and potassium chloride was significantly decreased. Contractions of rings from non-pregnant rats, pretreated with phenylephrine or potassium chloride, in response to calcium chloride were greater than those of similarly treated rings from pregnant rats. When the concentration of calcium chloride in the medium bathing the rings was reduced to 0.8 mmol·l^-1, the contractile response to phenylephrine was significantly (P<0.005) inhibited in rings from both pregnant and non-pregnant rats but to a greater extent in rings from non-pregnant rats. Contractions of aortic rings from pregnant rats in response to phenylephrine in calcium-free medium were similar to those of rings from non-pregnant rats, suggesting equal dependence on calcium from intracellular stores. The results suggest that pregnancy decreased the response to calcium influx into the aortic smooth muscle cells through both receptor-and voltage-operated calcium entry pathways. Since de-endothelialisation reversed the pregnancy-induced diminished contraction to phenylephrine, it is likely that pregnancy interferred with contractions induced by activation of receptors with phenylephrine through enhanced production of endothelium-derived relaxing factor(s).Abbreviations EC₅₀ concentration of drug producing 50% contraction - EDCF endothelium-derived contraction factor - EGTA ethyleneglycol-bis (-aminoethyl ether)-NN tetraacetic acid - PSS physiological salt solution - VSM vascular smooth muscle     

Morphology of the porcine myometrium during parturition

The ultrastructure of the porcine myometrium collected at well-defined stages during parturition was investigated by transmission electron microscopy. The morphology of the parturient myometrium resembled in general that in pregnant and non-pregnant pigs. The diameter of the smooth muscle cells was, however, about twice that of non-pregnant myometrium. Thick myofilaments were numerous. The number of caveolae seemed to be higher in parturient compared with non-pregnant and pregnant cells. Gap junctions occurred richly and were large, while the intrinsic innervation was very scanty. To conclude, the endocrinological changes in the pig taking place just prior to parturition, are translated into morphological changes by stimulating the formation of uterine gap junctions. This provides low-resistance pathways between the muscle cells and activates the myometrium for the delivery process.     

Immunohistochemistry using an antibody to unphosphorylated connexin 43 to identify human myometrial interstitial cells

Background

Myometrial smooth myocytes contract as a result of electrical signalling via a process called excitation-contraction coupling. This process is understood in great detail at the cellular level but the generation and coordination of electrical signals throughout the myometrium are incompletely understood. Recent evidence concerning the vital role of interstitial cells of Cajal in tissue-level signalling in gastrointestinal tract, and the presence of similar cells in urinary tract smooth muscle may be relevant for future research into myometrial contractility but there remains a lack of evidence regarding these cells in the myometrium.

Methods

Single stain immunohistochemical and double stain immunofluorescence techniques visualised antibodies directed against total connexin 43, unphosphorylated connexin 43, KIT, alpha-SMA and prolyl 4-hydroxylase in myometrial biopsies from 26 women representing all stages of reproductive life.

Results

Myometrial smooth myocytes from term uterine biopsies expressed connexin 43 in a punctate pattern typical of gap junctions. However, on the boundaries of the smooth muscle bundles, cells were present with a more uniform staining pattern. These cells continued to possess the same staining characteristics in non-pregnant biopsies whereas the smooth myocytes no longer expressed connexin 43. Immunohistochemistry using an antibody directed against connexin 43 unphosphorylated at serine 368 showed that it is this isoform that is expressed continually by these cells. Double-stain immunofluorescence for unphosphorylated connexin 43 and KIT, an established marker for interstitial cells, revealed a complete match indicating these cells are myometrial interstitial cells (MICs). MICs had elongated cell processes and were located mainly on the surface of the smooth muscle bundles and within the fibromuscular septum. No particular arrangement of cells as plexuses was observed. Antibody to prolyl 4-hydroxylase identified fibroblasts as separate from MICs.

Conclusion

MICs are identified consistently on the boundaries of smooth muscle bundles in both the pregnant and non-pregnant uterus and are distinct from fibroblasts. The uniform distribution of connexin 43 on the cell membrane of MICs, rather than localisation in gap junction plaques, may represent the presence of connexin hemichannels. This antibody specificity may aid future study of this potentially important cell type.     

Expression of cystathionine β-synthase and cystathionine γ-lyase in human pregnant myometrium and their roles in the control of uterine contractility

Background

Human uterus undergoes distinct molecular and functional changes during pregnancy and parturition. Hydrogen sulfide (H₂S) has recently been shown to play a key role in the control of smooth muscle tension. The role of endogenous H₂S produced locally in the control of uterine contractility during labour is unknown.

Methodology/Principal Findings

Human myometrium biopsies were obtained from pregnant women undergoing cesarean section at term. Immunohistochemistry analysis showed that cystathionine-γ-lyase (CSE) and cystathionine-β-synthetase (CBS), the principle enzymes responsible for H₂S generation, were mainly localized to smooth muscle cells of human pregnant myometrium. The mRNA and protein expression of CBS as well as H₂S production rate were down-regulated in labouring tissues compared to nonlabouring tissues. Cumulative administration of L-cysteine (10⁻⁷–10⁻² mol/L), a precursor of H₂S, caused a dose-dependent decrease in the amplitude of spontaneous contractions in nonlabouring and labouring myometrium strips. L-cysteine at high concentration (10⁻³ mol/L) increased the frequency of spontaneous contractions and induced tonic contraction. These effects of L-cysteine were blocked by the inhibitors of CBS and CSE. Pre-treatment of myometrium strips with glibenclamide, an inhibitor of ATP-sensitive potassium (K_ATP) channels, abolished the inhibitory effect of L-cysteine on spontaneous contraction amplitude. The effects of L-cysteine on the amplitude of spontaneous contractions and baseline muscle tone were less potent in labouring tissues than that in nonlabouring strips.

Conclusion/Significance

H₂S generated by CSE and CBS locally exerts dual effects on the contractility of pregnant myometrium. Expression of H₂S synthetic enzymes is down-regulated during labour, suggesting that H₂S is one of the factors involved in the transition of pregnant uterus from quiescence to contractile state after onset of parturition.     

Distribution of estrogen and progesteron receptors in the uterus: an immunohistochemical study in the immature and adult pseudopregnant rabbit

Summary In order to clarify the distribution and content of estrogen (ER) and progesteron receptors (PR) under changing hormonal influences within the various cell populations of the uterus (glandular and luminal endometrial epithelium, stroma, myometrium), immunohistochemical determinations using specific monoclonal antibodies were made. To correlate the immunohistochemical findings with peripheral hormone levels and specific tasks of the endometrium, 17-estradiol and progesterone serum levels were measured and cell proliferation determined by use of BrdU-labelling-immunohistochemistry. At the subcellular level ER and PR were located exclusively in the cell nuclei of female rabbits, which were either immature and lacking any peripheral hormone levels or were pseudopregnant (d0–d8 p.hCG). In the immature rabbits a general faint ER and PR immunostaining was found. In addition to a general increase in ER and PR in all cell populations estrous rabbits (d0 p.hCG) showed a significant rise of ER in the epithelial cells and of PR in the myometrium. Within the epithelial cells and the myometrium the ER dropped heavily within a few days of pseudopregnancy. The PR, however, increased sharply during the first two days of pseudopregnancy and decreased gradually following d4 p.hCG. A close relationship was observed between the high PR content and the proliferation rate of the epithelial cells on d2 p.hCG. In spite of the more rapid decrease of ER compared with PR, the glandular epithelium retained positive immunostaining. In the stroma the ER and especially PR content did not change significantly during the course of pseudopregnancy suggesting that some of the well-known differentiation events in the luminal epithelium may be mediated by the stroma.     

Binding of 125I-epidermal growth factor in human uterus

Summary Quantitative light-microscopic autoradiography was used on five human uteri at two different phases of the menstrual cycle to ascertain the cell types with binding sites for epidermal growth factor (EGF). The results revealed that stromal cells, glandular epithelium of endometrium, elongated and circular muscle cells of myometrium, smooth muscle and endothelial cells of arterioles in the basal endometrium and myometrium contained numerous silver grains following incubation with ¹²⁵I-EGF. Coincubation with 100-fold excess unlabeled EGF resulted in a complete disappearance of silver grains from all cell types. Quantitative grain analysis indicated that stromal cells contained the highest number of EGF-binding sites (P<0.05) with no significant differences among the others (P>0.05). There was no cyclic variation of EGF-binding to any of the uterine cell types. The present data demonstrate that all the cell types of human uterus, including arterioles, contain EGF-binding sites. This suggests that all the cells in human uterus subserving different functions are targets of EGF action.     

The innervation of the myometrium of the cynomolgus monkey (Macaca fascicularis)

Summary The autonomic innervation of the myometrium of Macaca fascicularis consists of bundles of unmyelinated nerve fibres running between the smooth muscle cells, and is therefore considered to be of the fascicular (= unitary) type. Close contacts between nerve fibres and smooth muscle fibres were not found. Modification of the chromaffin method according to Tranzer and Richards made it possible to visualize the heterogeneity of the nerve fibres in a single bundle. The following fibre types were found to coexist: (1) noradrenergic fibres containing synaptic vesicles with a dense granule, (2) cholinergic fibres containing empty synaptic vesicles, and (3) non-adrenergic noncholinergic (NANC) fibres containing only or predominantly large dense-cored vesicles, which do not react with this method. Noradrenergic fibres are the most numerous (around 60%), followed by NANC fibres (30%) and cholinergic elements (around 10%). The distribution of these three types is similar in the cervix, the isthmus and the body of the uterus in pregnant and non-pregnant females.     

Immunocytochemical analysis of oestrogen receptors and progesterone receptors in the human uterus throughout the menstrual cycle and after the menopause.

To obtain more insight into the relationship between cyclic and regional changes in steroid receptor expression and function-related changes in the various types of cell of the normal human uterus, we performed an immunocytochemical study on paraffin-embedded sections. The distribution and intensity of immunostaining for the oestrogen receptor and the progesterone receptor in the various types of cell were semiquantitatively scored. The data were statistically compared for the different phases of the menstrual cycle and after the menopause, and for the different regions of the corpus and (endo)cervix uteri. During the menstrual cycle, significant changes in oestrogen receptor score were observed in glandular and stromal cells of endometrium basalis and functionalis and in smooth muscle cells of the myometrium. In all types of cell, oestrogen receptor expression reached a maximum in the late proliferative phase. During the early secretory phase, oestrogen receptor staining declined sharply in stromal and smooth muscle cells, whereas, in glandular epithelium, oestrogen receptor expression decreased more gradually. During mid- and late-secretory phases, an increase in oestrogen receptor staining was also observed in predecidualizing stromal cells and smooth muscle cells. Progesterone receptor numbers changed significantly in glandular epithelium but not in stromal and smooth muscle cells. Glandular progesterone receptor expression reached a maximum in the early secretory phase and was then drastically reduced. During mid- and late-secretory phases stromal cells were moderately stained for progesterone receptor in contrast to epithelial gland cells which showed no or very weak staining. No regional variations in steroid receptor distribution in endometrium and myometrium were found.(ABSTRACT TRUNCATED AT 250 WORDS)     

Immunohistochemical localization of constitutive and inducible cyclo-oxygenases in rat uterus during the oestrous cycle and pregnancy

The uterus is a rich source of eicosanoids synthesized from arachidonic acid metabolism through the cyclo-oxygenase pathway. Two isoforms of cyclo-oxygenase, constitutive (COX-I) and inducible (COX-II) enzyme, have been reported. In the present study, we have immunohistochemically mapped the distribution of both COX-I and COX-II during various physiological states of the rat uterus. Uterine tissue was collected from female rats (a) during different stages of the oestrous cycle, (b) on days 1, 4, 8 and 18 of gestation, (c) after spontaneous delivery and (d) post partum, and fixed in Bouin's fixative. After paraffin wax embedding, 5-m-thick sections were immunohistochemically stained by the ABC technique. Observation of the stained sections under the light microscope revealed that, in non-pregnant rat uterus, both COX-I and COX-II were abundantly expressed in the endometrium, with minimal staining observed in the myometrium. Staining was more prominent in epithelial cells than in stromal cells. The intensity of staining in epithelial cells was highest at pro-oestrus and oestrus and lowest at dioestrus. In pregnant rats, although the expression of both COX-I and COX-II was localized primarily to the endometrium with very little staining in the myometrium on day 1 of gestation, both of these enzymes were also apparent in myometrial cells by day 4 of gestation. The staining intensity of endometrial and myometrial cells increased further with the progression of gestation, being maximal at the time of spontaneous delivery. During the post-partum period, however, the staining intensity for both of the enzymes in endometrium and myometrium was decreased. Thus, our studies show that the expression of cyclo-oxygenases in various uterine cells vary with the oestrous cycle and with pregnancy. Furthermore, prominent increases in the expression of cyclo-oxygenases in the myometrium during pregnancy and parturition imply that the cyclo-oxygenase system in the myometrium may play a major role in modulating uterine contractility during pregnancy and labour. © 1998 Chapman & Hall     

Expression of nidogens in rat uterus and embryo during decidualization and implantation

The purpose of this study was to demonstrate the expression of nidogen-1 and nidogen-2 and their possible role in decidualization and implantation events during early pregnancy in rats. The tissue samples were examined from pregnant animals between gestational days 1-8 using immunocytochemistry. The uterine luminal epithelium, the glandular epithelium, and the myometrial smooth muscle cells stained strongly from gestational days 1-8 with both nidogen antibodies. At day 4 the decidual reaction areas began to appear in the stromal matrix and immunostaining of both nidogens revealed that the basement membrane of the surface epithelium was discontinuous. The differentiation of stromal cells into decidual cells was seen at gestational day 5 and both nidogens were weakly expressed in the decidualizing cells. At day 6, nidogen-2 immunoreactivity was higher in the primary decidual cells close to the embryo than nidogen-1, and during development of the decidual tissue both nidogens appeared in the endometrial stromal cells. At day 7, while expression of both nidogens declined in the primary decidual cells, their expression was markedly observed in the secondary decidual cells close to the myometrium. At day 8, expression of both nidogens was also observed to increase in the primary decidual cells. While nidogen-2 expression was seen in the parietal endoderm and primary ectoderm of the rat embryos at this developmental stage, nidogen-1 expression was only detected in the parietal endoderm. These results indicate that nidogen-1 and nidogen-2 could play important roles during embryogenesis, decidualization, and implantation in the endometrium of rat uterus.     

Decreased smooth muscle side effects with 16, 16-dimethyl-trans-Δ2-PGE1 methyl ester in Japanese monkey (Macaca fuscata fuscata)

The smooth muscle stimulating activity of a new PGE₁ analog, 16, 16-dimethyl-trans-Δ²-PGE₁ methyl ester (ONO-802) was evaluated by simulataneous;y recording the EMG of the uterus and intestines, along with urinary bladder pressure, and blood pressure in pregnant and non-pregnant Japanese monkeys ($\text{Macaca fuscata fuscata}$). Single intravenous injections of ONO-802 in increasing dosages (0.2–5 μg/kg) were found to be 50–100 times or more effective in inducing uterine contraction than PGF₂α and PGE₁. A mild, transient gastrointestinal muscle stimulating activity was observed, but change in urinary bladder pressure and blood pressure was not evident. ONO-802 induced uterine contractions in the pregnant animals were 10 times greater than in the non-pregnant animals. These results suggest that ONO-802 may be a suitable clinical prostaglandin for use in therapeutic abortion.     

Progesterone-dependent expression of keratinocyte growth factor mRNA in stromal cells of the primate endometrium: keratinocyte growth factor as a progestomedin

In vitro studies have shown that keratinocyte growth factor (KGF, also known as FGF-7) is secreted by fibroblasts and is mitogenic specifically for epithelial cells. Therefore, KGF may be an important paracrine mediator of epithelial cell proliferation in vivo. Because stromal cells are thought to influence glandular proliferation in the primate endometrium, we investigated the hormonal regulation and cellular localization of KGF mRNA expression in the rhesus monkey uterus. Tissues were obtained both from naturally cycling monkeys in the follicular and luteal phases of the cycle, and from spayed monkeys that were either untreated or treated with estradiol (E2) alone, E2 followed by progesterone (P), E2 plus P, or E2 plus P plus an antiprogestin (RU 486). Northern blot analysis of total RNA with 32P- labeled probes revealed that the level of KGF mRNA in the endometrium was 70-100-fold greater in the luteal phase or after P treatment than in untreated, E2-treated, or follicular phase animals. Northern analysis also showed that KGF mRNA was present in the myometrium but was unaffected by hormonal state. RU 486 treatment prevented the P- induced elevation of endometrial KGF mRNA. P-dependent elevation of endometrial KGF expression was confirmed by measurement of KGF protein in tissue extracts using a two-site enzyme-linked immunosorbent assay. In situ hybridization with nonradioactive digoxigenin-labeled cDNA probes revealed that the KGF mRNA signal, which was present only in stromal and smooth muscle cells, was substantially increased by P primarily in the stromal cells located in the basalis region. Smooth muscle cells in the myometrium and the walls of the spiral arteries also expressed KGF mRNA, but the degree of this expression did not differ with hormonal state. P treatment led to increased proliferation in the glandular epithelium of the basalis region and to extensive growth of the spiral arteries. We conclude that the P-dependent increase in endometrial KGF resulted from a dual action of P: (a) a P- dependent induction of KGF expression in stromal cells, especially those in the basalis (zones III and IV), and (b) a P-dependent increase in the number of KGF-positive vascular smooth muscle cells caused by the proliferation of the spiral arteries. KGF is one of the first examples in primates of a P-induced, stromally derived growth factor that might function as a progestomedin.     

Increase in concentrations of prostaglandin endoperoxide synthase and prostacyclin synthase in human myometrium in late pregnancy

Concentrations of prostaglandin endoperoxide synthase (i.e. cyclooxygenase; PGH synthase) and prostacyclin synthase (PGI synthase) were quantified with specific radioimmunometric assays inhuman myometrium during the last trimester of pregnancy (n=23) and in non-pregnant controls (n=8). Pregnant myometrium contained 3 times more PGH synthase per mg microsomal protein than non-pregnant myometrium (p < 0.01) but there was no increase with increasing gestational age in the third trimester nor with the onset of labor. In pregnancy, as compared to the non-pregnant state, there was no significant change in the PGI synthase content of myometrial microsomes, but significantly more PGI synthase was recovered in other subcellular fractions (p < 0.01). This suggests that pregnancy affects preferential changes in the subcellular distribution of PGI synthase in myometrial cells.Relative to its PGI synthase content pregnant myometrium contained twice as much PGH synthase as non-pregnant myometrium (p < 0.01). This may offer further evidence that PGH synthase rather than PGI synthase itself is the rate limiting factor in myometrial PGI₂ production. On the other hand, the much larger increase in PGH synthase than in PGI synthase in pregnant as compared to non-pregnant myometrium, may serve to promote preferential synthesis of prostaglandins that are potent myometrial stimulants and of critical importance in human parturition.     

Platelet-activating factor (PAF)-like activity, localization of PAF receptor (PAF-R) and PAF-acetylhydrolase (PAF-AH) activity in bovine endometrium at different stages of the estrous cycle and early pregnancy

PAF-like activity in the endometrium increased from days 2-4 to day 12 and day 20 in both cyclic and pregnant cows. There was an increase in platelet aggregation induced by PAF-like activity in the endometrium of pregnant animals on day 20 as compared to cyclic animals at the same point in time. Two major bands of PAF-R protein at 67 kDa and 97 kDa were detected by Western blot analysis. PAF-R was localized mainly in luminal and glandular epithelium of the endometrium, but the staining was markedly increased in the endometrium of pregnant cows on day 20 compared to cyclic animals on the same day. The purified PAF-AH from the endometrium is similar to in plasma. In cyclic cattle, no changes in PAF-AH activity of endometrium were observed, whereas a decrease in enzyme activity occurred in pregnant cows on day 20 as compared to cyclic animals on the same day. We suggest that the bovine endometrium produces PAF-like activity, expresses the PAF-R and possesses a PAF-AH activity which varies during pregnancy.     

Platelet-activating factor (PAF)-like activity, localization of PAF receptor (PAF-R) and PAF-acetylhydrolase (PAF-AH) activity in bovine endometrium at different stages of the estrous cycle and early pregnancy*

PAF-like activity in the endometrium increased from days 2-4 to day 12 and day 20 in both cyclic and pregnant cows. There was an increase in platelet aggregation induced by PAF-like activity in the endometrium of pregnant animals on day 20 as compared to cyclic animals at the same point in time. Two major bands of PAF-R protein at 67 kDa and 97 kDa were detected by Western blot analysis. PAF-R was localized mainly in luminal and glandular epithelium of the endometrium, but the staining was markedly increased in the endometrium of pregnant cows on day 20 compared to cyclic animals on the same day. The purified PAF-AH from the endometrium is similar to in plasma. In cyclic cattle, no changes in PAF-AH activity of endometrium were observed, whereas a decrease in enzyme activity occurred in pregnant cows on day 20 as compared to cyclic animals on the same day. We suggest that the bovine endometrium produces PAF-like activity, expresses the PAF-R and possesses a PAF-AH activity which varies during pregnancy.     

Renin in the uterus of non-pregnant mice

Summary The distribution of renin in the uterus of non-pregnant mice was studied by immunocytochemistry, using the peroxidase-antiperoxidase (PAP) technique, by electron microscopy and biochemical renin estimation. By immunocytochemistry, two separate localizations of renin could be identified, one in the apical region of the endometrial cells, the other in perivascular cells of the myometrium. The renin concentration in the endometrium, as determined cytochemically, and the total renin content of the uterus, as estimated enzymatically, varied with the ovarian cycle, being highest during the oestrus and lowest during the post-oestrus phase. The electronmicroscopic finding of fluctuations in the granule content of the epithelial cells was in accordance with the immunocytochemical and biochemical data.These studies were supported by the Deutsche Forschungsgemeinschaft with the SFB 90 Cardiovasculäres System     

Regulation of cGMP-induced relaxation and cGMP-dependent protein kinase in rat myometrium during pregnancy

Increases inguanosine 3',5'-cyclic monophosphate (cGMP) induced bynitric oxide (NO), nitrovasodilators, and atrial peptides correlatewith relaxation of vascular smooth muscle. Relaxation of myometrialsmooth muscle by increases in cGMP, however, has required unusuallyhigh concentrations of the cyclic nucleotide. We tested the hypothesisthat the sensitivity of myometrium to relaxation by cGMP is increasedduring pregnancy. Aortic smooth muscle was more sensitive to relaxationby cGMP than myometrial tissues, and, contrary to our hypothesis,myometrium from pregnant rats was least sensitive. Although levels ofcGMP were elevated after treatment with the NO donor,S-nitroso-N-acetylpenicillamine, relaxation of myometrial tissues obtained from pregnant rats occurred only at extraordinarily high concentrations. The levels ofcGMP-dependent protein kinase (PKG) were significantly decreased inmyometrium from pregnant rats compared with myometrium from nonpregnantcycling animals or aortic smooth muscle. Administration of estradiol to ovariectomized rats increased myometrial PKG expression, andprogesterone antagonized this response. We conclude that1) myometrial tissues from pregnantrats are not sensitive to relaxation by cGMP and 2) this insensitivity to cGMP isaccompanied by progesterone-mediated decreases in the level of PKGexpression.

     

Connexin43 in porcine myocardium and non-pregnant myometrium

It is generally accepted that connexin43 (Cx43) is a major constituent of heart and myometrial gap junctions. However, the presence of Cx43 gap junctions in non-pregnant myometrium is still poorly documented. Tissue sections of porcine heart and non-pregnant uterus and myometrial smooth muscle cell cultures were immunostained with monoclonal antibody against Cx43. In the heart, intensive immunostaining was confined to the intercalated discs as previously reported. In the non-pregnant uterus, punctuate immunostaining of Cx43 was seen throughout the myometrium along cell interfaces between myocytes. The expression of Cx43 was sustained in cultured smooth muscle cells isolated from non-pregnant myometrium. Western blotting has detected single isoform of Cx43 in both, cardiac and myometrial tissues. The electrophoretic mobility of porcine heart Cx43 was similar to that of myometrial isoform but different from the pattern of mobility of Cx43 of the rat heart. Hence, porcine myometrium may provide attractive model for studying cellular mechanisms triggering expression of gap junction protein in normal (non-pregnant) uterus.     

The effect of progestin on rabbit endometrial aromatase activity

The ability of rabbit endometrium to synthesize estrogen was investigated. Aromatase activity was measured by incubating the viable tissue fragments with [3H] testosterone at 37 degrees C and quantitated the [3H]estrogens at the end of 6 h incubation. The activity was not detectable in endometrium and myometrium of non-pregnant rabbits. In pregnant rabbit, the activity increased from 2 to 12 fmol/h/g of tissue in endometrium obtained from both gravid and nongravid uterine horns 4-8 days after mating. Aromatase activity in endometrium at implanted sites on the 8th day after mating reached up to 550 fmol/h/g of tissue. Hormonal regulation of the aromatase activity was studied directly in cultured rabbit endometrial stromal cells. Both progesterone and medroxyprogesterone acetate stimulated the activity with 2- to 25-fold increase over the control values (0.2-1 fmol/h/mg protein). The stimulation of aromatase activity by progestin was found to be both time- and dose-dependent. Estrogen, Bu2cAMP, forskolin, HCG or relaxin had no effect on the activity. Aromatase activity in glandular epithelial cells was neither detectable nor affected by progestin. These results indicate that aromatase activity in rabbit endometrium is concentrated in stromal cells and progestin stimulates the activity. However, estrogen, forskolin and relaxin which enhanced the stimulation of the activity by progestin in human endometrium had no effect on aromatase in rabbit endometrium. The increase of aromatase activity observed after the onset of conception may be physiologically important to increase the local estrogen content for decidualization of the endometrium.